예제 #1
0
 def test_vocab(self):
   vocab = vocabularies.ByteVocabulary()
   self.assertEqual(259, vocab.vocab_size)
   self.assertSequenceEqual(self.TEST_BYTE_IDS, vocab.encode(self.TEST_STRING))
   self.assertEqual(self.TEST_STRING, vocab.decode(self.TEST_BYTE_IDS))
   self.assertEqual(
       self.TEST_BYTE_IDS,
       tuple(vocab.encode_tf(self.TEST_STRING).numpy()))
   self.assertEqual(self.TEST_STRING, _decode_tf(vocab, self.TEST_BYTE_IDS))
예제 #2
0
 def test_not_equal(self):
     vocab1 = vocabularies.ByteVocabulary()
     vocab2 = vocabularies.ByteVocabulary(10)
     self.assertNotEqual(vocab1, vocab2)
예제 #3
0
 def test_out_of_vocab(self):
     vocab = vocabularies.ByteVocabulary()
     self.assertEqual(259, vocab.vocab_size)
     self.assertEqual("", vocab.decode([260]))
예제 #4
0
 def test_extra_ids(self):
     vocab = vocabularies.ByteVocabulary(extra_ids=10)
     self.assertEqual(269, vocab.vocab_size)
     self.assertEqual("a", vocab.decode([100]))
     self.assertEqual("", vocab.decode([268]))